Experimenting with Data Visualization in VR

By Katie Metzler

There’s considerable buzz around the promises of VR as a solution for data visualization. Granted, there is a lot of buzz around the promises of VR as a solution for everything. (My personal favourite is Air France’s provision of VR headsets as a solution for getting millennials onto your airline.) And for some in this space, the buzz is translating into serious cash. In February, Virtualitics raised $7 million in Series B funding to build “a sort of “Excel for VR,” that lets people see data in a more immersive and collaborative way”.  Depending on what you think of Excel, this is either a very good or a very bad thing.

But is the excitement warranted? Is looking at data in 3D or in VR really that much better than the 2D experience most of us are used to? As Lucas Matney of Techcrunch puts it, “it’s a little unclear whether the jump from 2D to 3D is really giving users a ton of extra insights...there has seemed to be a bit of overselling when it comes to how great AR/VR is for looking at stuff more immersively.”

To learn more about the potential for VR to transform data visualization and to try to answer some of these questions for ourselves, I worked with the team at Datavized, a startup based in New York City, and my SAGE colleagues, Diana Aleman and Andrew Boney, on a project to turn data from one of SAGE’s data products, SAGE Stats, into a 3D VR experience.

About the project

Our aim with this experiment was to find an inexpensive and quick way to try out a data visualization in VR. We wanted to learn more about the companies out there working on this, and to work through what some of the use cases might be for SAGE, an academic publisher with a long history of research methods publishing and growing list of data viz books and online courses. We had two initial ideas - the first was to produce an experience that would act as a delighter at conferences where we were meeting with librarians about SAGE Stats and wanted to create a fun and memorable experience in the booth that they could link to the product. The second potential use case was to incorporate 3D visualizations that could be viewed in VR within the product itself, as a way to enhance the map-based data visualizations already present within the tool and to highlight the story behind the data. In the end, we went with the first use case for this experiment, though we’re still excited by the idea of the second in the longer-term.

We were attracted to Datavized product, GeoMetric, because it allowed for 3D visualisation of data on a globe, which matched up with the kind of geospatial data series which are included in SAGE Stats.

About Datavized

Datavized Technologies is a software development company specialized in data visualization and WebXR solutions. Datavized immersive visualization software products include tools that make it easy to turn complex data into fully interactive audio/visual web experiences on all connected devices; including desktop, mobile, tablet and Virtual Reality (VR), Augmented Reality (AR) and Mixed Reality (MR) headsets, enabling users to tell immersive data-driven stories without having to code. For further information on Datavized, visit datavized.com.

About the GeoMetric tool

GeoMetric powered by Datavized is a web-based drag and drop dashboard tool that turns spreadsheets with location data into fully interactive 3D experiences. The 360-degree virtual maps can be easily accessed via a web link on mobile, tablet and desktop (no headsets needed) and seamlessly integrate with existing websites. GeoMetric makes it easy to run rich 3D geospatial data as fully interactive visualizations live in the browser and add spreadsheet data for analysis and presentation. With GeoMetric, non-technical users can easily turn CSV, XLSX and geographic data structures into fully interactive data visualizations. To request beta access or a product demo, click here or visit datavized.com.

What we built

Using GeoMetric in tandem with a data series from SAGE Stats on Persons Not Covered by Health Insurance, we created a 3D visualization which showed how the number of uninsured people in each state across the United States changed over time.

One of the stories we wanted to show through this data was how the introduction of Obamacare had a significant positive impact on the number of people without insurance, which can be seen when comparing data from 2013 to data from 2014, the year the Affordable Care Act went into effect.

 GeoMetric powered by Datavized - ‘Persons Not Covered by Health Insurance’ Source: SAGE Stats

GeoMetric powered by Datavized - ‘Persons Not Covered by Health Insurance’ Source: SAGE Stats

 GeoMetric powered by Datavized - ‘Persons Not Covered by Health Insurance’ Source: SAGE Stats

GeoMetric powered by Datavized - ‘Persons Not Covered by Health Insurance’ Source: SAGE Stats

What we learned

From this experiment, we took away a couple of key learnings about the value of VR for data visualization, and also some ideas for what could have made the experience even better.

Lesson 1: VR is fun!

Before rolling out the demo to potential SAGE Stats customers, we tried it out on our colleagues, and the takeaway from this was that trying out anything in VR is pretty fun. Watching other people trying out VR is also fun. There was lots of laughter and some “ooohing” and “ahhing”, and that’s what you want to hear if you’re trying to create a buzz at your conference stall.

KM.png

Lesson 2: Be clear about the story you’re trying to tell

The visualization we created was quite simple - we used a single data series from SAGE Stats and showed that plotted on a map of the US. We used the GeoMetric slider feature to show the change over time in the absolute number of people uninsured per state as well as the percentage. Unlike some of the use cases for VR for big data where the aim is to actually make analysis possible in a way that’s difficult in 2D because of the volume and complexity of the data, we weren’t trying to solve a problem with analytics, but to help the story of the dataset come through and encourage people to engage with the data.

As the implementation of Obamacare in 2014 was the major “story” in the data, on reflection, we should have done more to pull that story out in the demo, for example by labelling the introduction of Obamacare on the slider as a prompt for users to think about and compare the data before and after this event.

Lesson 3: Multi-platform is key

One of the key features of Datavized’s GeoMetric products is that they are browser-based and so work across a range of devices and can be experienced as 3D visualizations within your browser in addition to offering the option to toggle to VR. Datavized offered excellent support in helping SAGE to decide which VR equipment to purchase. We went for a Vive set up, as we’re keen to continue to experiment with producing VR experiences and so the investment in kit seemed worthwhile.

Setting up the Vive kit for the first time for a demo with colleagues was really simple and took about half an hour. The second time I demoed the experience, it was in the US and we rented the equipment and it was annoyingly difficult to get the sensors to connect and I ended up fiddling around with it for ages. Luckily, I got it working in the end, but even if I hadn’t been able to, I could have shown the visualization in the browser and that would have been a life saver.

Lesson 4:  Expectations of VR are very high

For many of my colleagues, this was their first experience of immersive VR. Despite this, their expectations of what they’d be able to do was very high. A number of people trying out what we’d built wanted to be able to dig in further to see state level data, and then “look around” that state, as if they were in Google Maps and switching to Street View. In general, there was an anticipation of a more game-like experience, even though most of them had never gamed in VR. This can be an issue if your aim is to create a fun and memorable experience, and people leave disappointed because they have very high expectations. I think the lesson here is just to be mindful of this when thinking about how to create a memorable, positive experience which helps people to connect with your product. It may be worth thinking about how to prepare people a little for the experience before they put on the headset so that they know what to expect.

Summary

In summary, working with Datavized was great and we’ve produced a cool little experience to show off SAGE Stats at conferences. We’ll be trying it out with librarians at the ALA annual conference in New Orleans this week! Pop by and try it out at stand #1549.

Want to learn more about data visualization? Join our upcoming SAGE Campus course starting on the 25th of June with Andy Kirk. Use the discount code DSJUNE25 to get 25% off.

Want to try out SAGE Stats? Sign up for a free trial here.

About

Katie Metzler is Executive Head of Methods Innovation and part of the SAGE Ocean team. SAGE Ocean’s mission is to improve social science by giving every researcher the skills and tools they need to work with big data and new tech. We’d love to hear from you so please do get involved! Follow Katie on Twitter at @KMetzlerSAGE.